Relationships and taxonomic status of Alternaria radicina ...

Mycologia, 94(1), 2002, pp. 49?61. 2002 by The Mycological Society of America, Lawrence, KS 66044-8897

Relationships and taxonomic status of Alternaria radicina, A. carotiincultae, and A. petroselini based upon morphological, biochemical, and molecular characteristics

Barry M. Pryor1 Robert L. Gilbertson

Department of Plant Pathology, University of California, One Shields Avenue, Davis, California 95616

Abstract: Alternaria radicina, A. carotiincultae, and A. petroselini are closely related pathogens of umbelliferous crops. Relationships among these fungi were determined based on growth rate, spore morphology, cultural characteristics, toxin production, and host range. Random amplified polymorphic DNA (RAPD) analysis of these species, other species of Alternaria, and closely related fungi was also performed. A. petroselini was readily differentiated from A. radicina and A. carotiincultae on the basis of spore morphology, production of microsclerotia, host range, and RAPD analysis. Alternaria radicina and A. carotiincultae were considerably more similar to each other than to A. petroselini, but could be differentiated on the basis of growth rate, spore morphology, colony morphology, and, to a limited extent, RAPD analysis. When grown on media having a high nutritional content, A. radicina produced a diffusible yellow pigment and crystals of the fungal metabolite radicinin. In contrast, A. carotiincultae produced little or no radicinin. However, when A. carotiincultae was grown on the same medium amended with radicinin, growth rate and colony and conidial morphology were more similar to those of A. radicina. These results suggest that the morphological differences between A. radicina and A. carotiincultae are due, at least in part, to radicinin production, and that these fungi are conspecific. Therefore, we propose that A. carotiincultae be considered a synonym of A. radicina.

Key Words: Apiaceae, carrot black rot, radicinin, rDNA

INTRODUCTION

Alternaria radicina Meier, Drechsler and E. D. Eddy is the causal agent of carrot black rot disease, which causes carrot crop losses worldwide (CAB 1972,

Accepted for publication May 11, 2001. 1 Corresponding author, phone: 520-626-5312, Fax: 520-621-9290, Email: bmpryor@ag.arizona.edu

Meier et al 1922). The fungus attacks all parts of the carrot plant including leaves, flowers, seeds, and roots. However, it is the necrotic black lesions induced in storage roots that result in the greatest economic losses (Grogan and Snyder 1952, Lauritzen 1926, Maude 1966). The host range of A. radicina is limited primarily to carrot. However, A. radicina also has been reported to cause a foliar blight of parsley and a stalk and root rot of celery (Tahvonen 1978, Wearing 1980).

Neergaard (1945) examined the morphological and cultural variation within the species A. radicina [as Stemphylium radicinum (Meier, Drechs. and E. D. Eddy) Neergaard 1939], and recognized three distinct types (a, b, and c) based on differences in colony morphology and, to a lesser extent, spore morphology and host range. Type a and b isolates were isolated from and were strongly pathogenic on carrot, whereas type c isolates were isolated from celery and were strongly pathogenic on celery and parsley. Neergaard (1945) further concluded that another closely related species, A. petroselini (Neergaard) Simmons (as S. petroselini Neergaard 1942), which is pathogenic on parsley, should be reclassified as S. radicinum var. petroselini Neergaard due to similarities in conidium morphology with S. radicinum.

The most significant differences between A. radicina type a and b isolates were in terms of their cultural characteristics (Neergaard 1945). On malt agar, type a isolates formed small colonies with irregular margins and produced tree-like crystals that grew into the medium. Type b isolates formed large colonies with even margins and usually did not produce crystals. The crystals produced by type a isolates were subsequently shown to be composed of radicinin, a keto-lactone secondary metabolite that has phytotoxic and antifungal properties (Aldridge and Grove 1964, Grove 1964, 1970). Radicinin also is produced by Alternaria chrysanthami, Cochliobolus lunata, Bipolaris coicis, and Phoma andina (Nakajima et al 1997, Noordeloos et al 1993, Nukina and Marumo 1977, Robeson et al 1982).

During the course of our research on carrot black rot disease in California, over 200 isolates of A. radicina were recovered from commercial carrot seed, carrot tissue, and soil. Most of these isolates (98%) corresponded to Neergaard's type a, i.e., slow growth,

49

50

MYCOLOGIA

irregular colony margin, and crystal production on potato dextrose agar (PDA). These isolates also produce a diffusible yellow pigment after 10?15 d on PDA or acidified PDA (APDA) (Pryor et al 1997). These isolates were designated as A. radicina type 1 by Pryor et al (1997). Six isolates, all recovered from carrot seed, had cultural characteristics that were more similar to those described by Neergaard as type b, i.e., rapid growth, even colony margin, and no crystal production. These latter isolates also did not produce a diffusible yellow pigment. These isolates were designated as A. radicina type 2 by Pryor et al (1997). Type 1 and 2 isolates were pathogenic on carrot (Pryor et al 1997).

In 1995, A. carotiincultae E. G. Simmons was described as a new species of Alternaria isolated from wild carrot from Ohio (Simmons 1995). Though similar in many respects to A. radicina, A. carotiincultae was classified as a distinct species based on having: (i) greater average conidium length, (ii) fewer obovoid and subspherical conidia, and (iii) a greater frequency of conidia produced in chains of two, or less commonly, three. A morphological comparison of the ex-type culture of A. carotiincultae with our A. radicina type 2 isolates suggested to us that they were conspecific.

In recent work, phylogenetic relationships among Alternaria, Ulocladium, and Stemphylium were determined based on analyses of nuclear 18S and ITS/ 5.8S, and mitochondrial small subunit (mt SSU) ribosomal DNA (rDNA) sequences (Pryor and Gilbertson 2000). This work revealed that A. radicina and A. petroselini were closely related, but had considerable divergence in 18S, ITS/5.8S, and mt SSU sequences. In contrast, A. radicina and A. carotiincultae had identical 18S and mt SSU sequences, and their ITS/5.8S sequences differed by one nucleotide. The high level of rDNA sequence identity between A. radicina and A. carotiincultae exceeds that found among A. dauci, A. solani, and A. porri, which have been considered as formae speciales of a single species by some authors ( Joly 1964, Neergaard 1945). Because both A. radicina and A. carotiincultae cause similar disease symptoms in carrot and have nearly identical rDNA sequences, it is hypothesized that A. radicina and A. carotiincultae may be conspecific.

The purpose of this work was to further examine the relationship between A. radicina and A. carotiincultae in terms of morphological, biochemical, and molecular characteristics. Isolates of A. petroselini were included because of the close relationship of this species with A. radicina.

MATERIALS AND METHODS

Cultural and morphological characterization. Three isolates each of A. radicina, A. carotiincultae, and A. petroselini were

TABLE I. Fungal isolates used in this study

Species

Sourcea

Designation

Isolates used in morphological, pathogenic, biochemical,

and RAPDb analyses:

Alternaria carotiincultae BMP

21-21-13

BMP

21-21-15

BMP

21-21-16

A. petroselini

BMP

21-41-01

BMP

21-41-02

BMP

21-41-03

A. radicina

BMP

21-21-07

BMP

21-21-11

BMP

21-21-14

Atypical isolates:

A. radicina (ex-neotype) ATCC

6503

A. radicina (celery isolate) ATCC

58405

Additional isolates used in RAPD analyses:

A. arborescensc

ATCC

28329

A. brassicicola

EEB

2232

A. carotiincultae (ex-type) EGS

26-010

A. dauci

ATCC

36613

A. macrospora

DGG

Amsl

A. petroselini

EGS

09-159

A. porri

ATCC

58175

A. radicina (representa- ATCC

96831

tive isolate)

A. solani

ATCC

58177

A. tenuissima

ATCC

16423

Stemphylium botryosum

ATCC

42170

S. callistephi

EEB

1055

S. sarcinaeforme

EEB

1072

S. vesicarium

ATCC

18521

a Abbreviations for sources are as follows: ATCC - American Type Culture Collection, Rockville, MD 20852; BMP B. M. Pryor, Dept. of Plant Pathology, University of California, Davis, CA 95616; DGG - D. G. Gilchrist, Dept. of Plant Pathology, University of California, Davis, CA 95616; EEB E. E. Butler, Dept. of Plant Pathology, University of California, Davis, CA 95616; EGS - E. G. Simmons, Crawfordsville, IN 47933.

b Random Amplified Polymorphic DNA. c A. alternata f. sp. lycopersici (Simmons, E. G. 1999. Mycotaxon 70:325?369).

used to determine growth rate, conidium dimensions, and cultural characteristics of these species (TABLE I). Alternaria radicina and A. carotiincultae were isolated from infested carrot seed, and A. petroselini isolates were obtained from infested parsley seed. The species identification of these isolates was accomplished by comparison of morphological and cultural characteristics with those of ex-type or representative cultures for each species [A. radicina, ATCC (American Type Culture Collection) 96831; A. carotiincultae, EGS (E. G. Simmons) 26?010 (ex-type); and A. petroselini, EGS 09?159]. The species identity for most of these isolates was confirmed by E. G. Simmons (717 Thornwood Rd., Crawfordsville, Indiana 47933). Isolates BMP 21?21?07

PRYOR AND GILBERTSON: TAXONOMIC STATUS OF ALTERNARIA CAROTIINCULTAE

51

and BMP 21?21?11 were identified as members of the radicina species-group, but were not confirmed as either A. carotiincultae or A. radicina. Isolates 21?41?01 and 21?41? 02 were not examined by E. G. Simmons. All isolates were maintained on PDA (Difco, Plymouth, Minnesota) plates at 22 C, or on PDA slants at 10 C.

Fungal growth rates were determined on potato carrot agar (PCA) (Simmons 1992), V-8 juice agar (V8A) (Simmons 1992), PDA, and acidified PDA (APDA, pH 5.0). For each isolate, three mm agar plugs were removed from the margin of a 10-day-old culture on corn meal agar (CMA; Difco) and placed, mycelium side up, in the center of each of three 9-cm Petri dishes containing the appropriate medium. Petri dishes were incubated in clear plastic boxes with the surface of each dish 40 cm beneath fluorescent lights (Sylvania cool-white, 10/14 h light/dark) at 22 C. Colony diameters were measured after 8 d of incubation. Values for the three plates were averaged to obtain a mean radial growth rate for each isolate. This experiment was conducted twice. For each medium, Bartlett's test for non-homogeneity was used to evaluate homogeneity of variance within each experiment. A chi-square test for homogeneity of variance was used to determine if data from the separate trials could be combined for a single analysis. An analysis of variance (ANOVA) was performed to determine significant differences (P 0.05) in growth rate among the three species for each medium.

Conidia from 10-day-old cultures on PCA, V8A, PDA, and APDA dishes, grown as described previously, were used to obtain conidium measurements. Conidia were taken from approximately 15 mm inside the colony margin in order to obtain uniform, mature conidia from an actively growing portion of the colony. Conidia were suspended in water and observed with a light microscope at 640. Length and width measurements were taken from three categories of conidia: those having 3, 4, or 5 transepta. Irregular, subspherical conidia with oblique septa, as well as immature conidia (i.e., lacking pigmentation and longisepta) were not counted. Measurements were taken from ten randomly selected conidia for each category. Mean values and length/width (l/w) ratios were calculated. In addition, the number of transepta per conidium was determined for 50 randomly selected conidia in four fields of view (200), and a mean number of septa per conidium were calculated. The maximum number of transepta per conidium was also noted for each species. For each medium, Bartlett's test for non-homogeneity was used to evaluate homogeneity of variance among isolates. ANOVA was performed to determine significant differences (P 0.05) in conidium length, width, and l/w ratio for each conidium category, and in the number of transepta/conidium among the three species for each medium.

Cultural characteristics were noted for each isolate after 10?15 d of growth on PCA, V8A, PDA, and APDA dishes. The production of pigments, crystals, and microsclerotia were determined by visual examination of dishes. The production of conidia in catenate arrangement was determined by examining dishes using a dissecting microscope (100) with substage illumination. The production of subspherical conidia was determined by examining a small section of

colony (approximately 1?2 mm2) mounted in water using a compound microscope (200).

Radicinin production. The capacity of each isolate to produce radicinin in liquid culture medium (20.7 g D-glucose, 1.2 g DL asparagine, 1.2 g K2HPO4?3H2O, 0.5 g MgSO4?7H2O, 0.5 g yeast extract, 0.1 g NaCl, per liter of deionized H2O) was determined. Inoculum for liquid cultures was prepared by flooding CMA plates containing 10day-old cultures of each fungal isolate with 10 mL of sterile deionized H2O, and dislodging most conidia and aerial mycelia with a plastic rod. For each isolate, two ml of this suspension was pipetted from the dish and added to 100 mL of liquid culture medium in a 500 mL Erlenmeyer flask. Flasks were placed on a rotary shaker at 120 rpm and incubated at 22 C for 10 d. After incubation, three mL of liquid medium were removed and extracted with chloroform as described by Robeson et al (1982). Radicinin was detected in chloroform extracts by thin-layer chromatography (TLC) analysis (Robeson et al 1982), and quantified by comparison with serial dilutions of reagent-grade radicinin (Sigma, St. Louis, Missouri) prepared in chloroform. This experiment was conducted three times.

The effect of radicinin on the growth of A. radicina and A. carotiincultae in culture was assessed by amending PDA with reagent-grade radicinin at concentrations of 100, 200, and 500 ppm. Three mm plugs with A. radicina (isolates 21?21?07 and ex-neotype ATCC 6503) or A. carotiincultae (ex-type EGS 26?010) were placed on dishes as previously described. After 10 d, the colony growth rate and average number of transepta per conidium was determined for each isolate on amended and non-amended media. This experiment was conducted three times. In addition, colony and conidium characteristics on amended and non-amended media were noted for each trial.

Pathogenicity tests. The host range of the isolates was determined by inoculating 5-wk-old carrot, dill, fennel, cilantro, and parsley seedlings; 6-week-old anise, caraway, and parsnip seedlings; and 8-wk-old celery seedlings. All seedlings had 3?5 true leaves. Seeds were sown in 10.5 cm-diam plastic pots and seedlings were thinned to 5 plants per pot. A suspension of conidia was prepared for each of the 9 test isolates by flooding a V-8 agar dish containing a 10-day-old culture with 10 mL of sterile deionized H2O and gently dislodging conidia with a plastic rod. Suspensions were filtered through two layers of cheesecloth, and the concentration of conidia was determined with a hemacytometer. Conidium suspensions were adjusted to 2000 conidia/mL, and sprayed onto leaves of each test plant, until run-off, with an aerosol spray bottle (Nalge Company, Rochester, New York). Two replicate pots of each host species were inoculated per isolate. One pot of uninoculated plants per host species was used as a control. Plastic bags were immediately placed over each pot, and secured at the base of each pot with rubber bands. After 48 h, the bags were removed and the pots were placed in a greenhouse mist chamber with periodic misting. Two wk after inoculation each plant was scored for the severity of disease with the following rating system: 0 no disease, 1 1% leaf necrosis, 2 5% leaf necrosis, 3 10% leaf necrosis, 4 20% leaf necrosis, 5 40% leaf

52

MYCOLOGIA

necrosis. Disease ratings for plants within each pot were averaged to obtain a mean rating for each pot. The mean ratings of two replicate pots were averaged to obtain a mean pathogenicity score for each fungal isolate on each host. The pathogenicity tests were conducted three times. Bartlett's test for non-homogeneity was used to determine if there was homogeneity of variance within each trial. A chisquare test for homogeneity of variance among trials was performed to determine if the data from the three trials could be combined into a single analysis. ANOVA was performed to determine significant differences (P 0.05) in the pathogenicity score among the three species for each host plant.

Molecular analysis. RAPD analyses were performed with isolates of A. radicina, A. carotiincultae, and A. petroselini listed in TABLE 1, as well as 11 additional isolates representing the genera Alternaria and Stemphylium (TABLE 1). All isolates were maintained on PDA plates at 22 C or on PDA slants at 10 C. DNA was extracted from fungi grown in liquid culture as previously described (Pryor and Gilbertson 2000), and further purified with the Prep-a-gene DNA Purification System (Biorad Inc., Hercules, California). DNA concentrations were adjusted to 10 ng/L for RAPD analyses.

RAPD analysis was performed using the 20 primers from Operon primer set A (Operon Technologies, Inc., Alameda, California). RAPD reactions were carried out in a 50 L reaction mixture (10 ng DNA, 0.5 mM each primer, 0.25 mM each dNTP, 2.5 mM MgCl2, and 1.0 U Amplitaq DNA polymerase in 1 Amplitaq PCR buffer II [PE Applied Biosystems, Foster City, California]). PCR was carried out in a thermal cycler (Model 480, PE Applied Biosystems) programmed for the following parameters: 94 C--1 min, 34 C--1.5 min, 72 C--2 min for 45 cycles. PCR products were analyzed by electrophoresis in 1% agarose gels in TBE buffer (Sambrook et al 1989), and UV illumination after staining in ethidium bromide. RAPD reactions were conducted at least twice with each primer to confirm that RAPD patterns were reproducible.

The sizes of RAPD fragments were determined with an IS-1000 Digital Imaging System (Alpha Innotech Corp., San Leandro, California). All RAPD fragments between 500 and 2500 base pairs (bp) were scored with no correction for band intensity. Isolates were scored for the presence or absence of a given RAPD marker, and a binary matrix was constructed. Cluster analysis of the data matrix was performed by the Unweighted Pair-Group Method with Arithmetic mean (UPGMA) using the NTSYS-pc Numerical Taxonomy and Multivariate Analysis System software (version 1.80, Exeter Software, Setauket, NY). Results were graphically displayed in a dendrogram.

Analysis of atypical A. radicina isolates. Two additional A. radicina isolates were examined during the course of this study: ATCC 6503, the ex-neotype of A. radicina (Simmons 1995), and ATCC 58405, an isolate from celery (Wearing 1980) (TABLE 1). Isolates were maintained on PDA plates at 22 C or on PDA slants at 10 C. Conidium measurements of each isolate were determined from cultures grown on PCA as previously described. Growth rate and cultural char-

FIG. 1. Radial growth of A. radicina, A. carotiincultae, and A. petroselini on various media. For each fungus, values represent averages of 3 isolates for three independent experiments.

acteristics, including production of pigments, crystals, subspherical conidia, microsclerotia, and/or catenate conidia were determined on APDA plates as previously described. Isolate ATCC 6503 was examined for the production of radicinin in liquid culture as described previously. The pathogenicity of ATCC 6503 and ATCC 58405 on carrot, caraway, celery, cilantro, dill, fennel, parsley, parsnip, and anise seedlings was determined as previously described. RAPD analyses of each isolate were conducted as described previously. In addition, the nuclear ITS/5.8S and mt SSU rDNA sequences were determined and compared to sequences of type/representative cultures in GenBank. For the molecular analyses, growth of fungal tissue, extraction of fungal DNA, sequence analyses, and sequence comparisons were performed as previously described (Pryor and Gilbertson 2000).

RESULTS

Cultural and morphological characterization. Statistical analysis revealed homogeneity of variance within and between the growth rate experiments for all 4 media tested. Thus, data from both experiments were combined into a single analysis. On all media, the growth of A. radicina was significantly less than that of A. carotiincultae or A. petroselini (FIG. 1), particularly on APDA where, in most cases, colony growth of the former ceased after 10?15 d. The growth of A. carotiincultae and A. petroselini was not significantly different on PCA or PDA, whereas on V8A the growth of A. petroselini was significantly less than that of A. carotiincultae, and on APDA the growth of A. carotiincultae was significantly less than that of A. petroselini.

Conidium measurements for the three species are presented in TABLE II. For each category of conidia, there was no significant difference in length, width, or l/w ratio of A. radicina and A. carotiincultae conidia. Conidia of A. petroselini were significantly wid-

PRYOR AND GILBERTSON: TAXONOMIC STATUS OF ALTERNARIA CAROTIINCULTAE

53

TABLE II. Conidium measurements for A. radicina, A. carotiincultae, and A. petroselini grown on various media

Mediumc

Species

Conidium measurements (l/ w/ l/w ratio)a

3-septa

4-septa

5-septa

# transverse septa/ conidiumb

Max

Avg

PCA

A. radicina

34a/ 16a/ 2.0b

41a/ 17a/ 2.4b

48a/ 18a/ 2.7b

7

4.0b

A. carotiincultae

35a/ 17a/ 2.1b

44ab/ 18a/ 2.4b 50a/ 18a/ 2.8b

10

4.7c

A. petroselini

37a/ 21b/ 1.8a

46b/ 22b/ 2.1a

54b/ 23b/ 2.3a

7

3.6a

V8A

A. radicina

38a/ 18a/ 2.1b

45a/ 19a/ 2.4a

50a/ 19a/ 2.6a

7

3.6a

A. carotiincultae

35a/ 17a/ 2.1b

46a/ 18a/ 2.5a

53a/ 19a/ 2.8a

10

4.6b

A. petroselini

39a/ 21b/ 1.9a

49a/ 22a/ 2.2a

54a/ 22a/ 2.5a

6

3.7a

PDA

A. radicina

37a/ 18a/ 2.1a

46a/ 19a/ 2.4b

52a/ 21a/ 2.5b

6

3.8a

A. carotiincultae

38a/ 18a/ 2.1a

47ab/ 20a/ 2.4b 55a/ 21a/ 2.6b

9

4.5b

A. petroselini

44b/ 22b/ 2.0a 50b/ 24b/ 2.1a 55a/ 24a/ 2.3a

6

3.6a

APDA

A. radicina

42a/ 19a/ 2.2a

47a/ 21a/ 2.2a

56ab/ 22b/ 2.5ab

6

3.3a

A. carotiincultae

38a/ 18a/ 2.1a

46a/ 19a/ 2.4a

54a/ 19a/ 2.8b

8

3.9b

A. petroselini

45a/ 23b/ 2.0a

53b/ 24b/ 2.2a

61b/ 25c/ 2.4a

5

3.1a

a Measurements were obtained from conidia collected from approximately 15 mm inside of the colony margin of a 10-dayold culture. Suspensions of conidia were observed with a compound microscope at 640X, and measurements were taken from randomly selected conidia having 3, 4, or 5 transepta. For each conidium category, ten conidia were measured and a mean value was calculated. For each medium, values within each column followed by different letters are significantly different (P 0.05).

b Fifty conidia from each isolate were randomly observed in four fields of view (200X) and the number of transverse septa per conidium was counted and the mean number of septa per conidium was calculated. For each medium, values within each column followed by different letters are significantly different (P 0.05).

c Abbreviations for media are as follows: PCA, potato-carrot agar; V8A, V8 juice agar; PDA, potato-dextrose agar; APDA, acidified PDA.

er and, in most cases, longer compared with conidia of A. radicina and A. carotiincultae. In most cases, l/ w ratios of A. petroselini conidia were significantly lower than those of A. radicina and A. carotiincultae.

Conidia of A. carotiincultae had significantly more transepta than those of A. radicina or A. petroselini (TABLE II). For example, on PCA A. carotiincultae conidia had a mean of 4.7 transepta, whereas A. radicina and A. petroselini conidia had means of 4.0 and 3.6 transepta, respectively. The maximum number of transepta per conidium was also consistently greater for A. carotiincultae compared with A. radicina or A. petroselini (TABLE II). The mean number of transepta per conidium for A. petroselini and A. radicina were not significantly different except on PCA.

Differences in colony morphology were observed, particularly on media with a high nutrient content (e.g., PDA and APDA) (TABLE III, FIG. 2). On such media, A. radicina colonies grew slowly and stopped growing after 10?15 d without covering the dish surface (FIG. 2). In addition, these colonies had irregular margins and produced dendritic crystals, characteristic of radicinin, and a yellow diffusible pigment (FIG. 3). In contrast, A. carotiincultae and A. petroselini colonies grew rapidly and covered the dish surface within 10?15 d, and these colonies had smooth margins (FIG. 2). Alternaria carotiincultae isolates did

not produce crystals or yellow pigments within 15 d (FIG. 2), although one isolate (21?21?13) produced very small crystals and a small amount of yellow pigment after 30 d of growth on APDA. Alternaria petroselini isolates produced few or no crystals, and trace amounts of yellow pigment on APDA.

All isolates of A. carotiincultae produced conidia in catenate arrangement on all media (TABLE III). Alternaria radicina isolates 21?21?07 and 21?21?11 commonly produced conidia in catenate arrangement on PCA, V8A, and PDA, but not on APDA. Production of conidia in catenate arrangement was uncommon for A. radicina isolate 21?21?14 on all media. The production of conidia in catenate arrangement by A. petroselini isolates was uncommon on PCA and V8A, and was not observed on PDA or APDA.

All isolates of A. radicina and A. petroselini produced subspherical conidia on all media (TABLE III). All isolates of A. carotiincultae commonly produced subspherical conidia on APDA but not on PCA, V8A, or PDA.

Alternaria petroselini isolates produced small microsclerotia (approximately 60?120 m in diameter) (TABLE III, FIG. 4), which were distributed throughout the medium beneath the agar surface. Production of microsclerotia occurred on all media but was greatest on V8A. On V8A, microsclerotia were first

54

MYCOLOGIA

TABLE III. Cultural characteristics for A. radicina, A. carotiincultae, and A. petroselini grown on various media

Mediuma

Characteristic

A. radicina

A. carotiincultae

A. petroselini

PCA V8A PDA APDA

rapid growthb catenate conidiac subspherical conidiad pigment productione crystal productionf microsclerotiag rapid growth catenate conidia subspherical conidia pigment production crystal production microsclerotia rapid growth catenate conidia subspherical conidia pigment production crystal production microsclerotia rapid growth catenate conidia subspherical conidia pigment production crystal production microsclerotia

/

/ / /

/ / / /

a Abbreviations for media are as follow: PCA, potato-carrot agar; V8A, V8 juice agar; PDA, potato-dextrose agar; APDA,

acidified PDA. b Rapid growth was recorded as () if the colony diameter was 7 cm after 10 days, or () if colony diameter was 6

cm after 10 days. c The presence of conidia in catenate arrangement was determined by examining dishes using a dissecting microscope

(100X) and substage illumination, and was recorded as common () if several could be viewed in a single field of view, uncommon (/) if it was necessary to scan several fields of view to find one example, or () if absent.

d The presence of subspherical conidia was determined by examining a small section of colony (approximately 1-2 mm2) mounted in water using a compound microscope (200X), and was recorded as common () if several such conidia could be viewed in a single field of view, uncommon (/) if it was necessary to scan several fields of view to find one example, or absent ().

e The production of pigment within 14 d was recorded as present (), faint (/), or absent (). f The production of crystals within 14 d was recorded as present () or absent (). g The presence or absence of microsclerotia, formed within the agar medium after 14 d of growth, was recorded as present () or absent ().

observed after 7 d of growth. Under the conditions of this study, A. radicina and A. carotiincultae did not produce microsclerotia.

Radicinin production. All isolates grew well and similarly in liquid culture medium based upon visual comparisons of total fungal biomass produced during incubation. When grown in liquid culture, A. radicina isolates produced radicinin, and TLC analyses revealed that isolates produced 300?700 mg radicinin/ L of medium. Alternaria carotiincultae isolates produced only trace amounts of radicinin (FIG. 5), consistent with the failure of these isolates to produce crystals in solid media. A. petroselini isolates produced more radicinin than A. carotiincultae, but consider-

ably less than A. radicina. Similar results were obtained in three independent experiments.

When A. carotiincultae was grown on PDA amended with radicinin (100 and 200 mg/l), colony growth after 10 d was less than that on non-amended medium, particularly at the 200 ppm rate (44 0.9 mm versus 64 0.7 mm, FIG. 6). Similarly, growth of A. radicina was reduced on PDA amended with radicinin compared with non-amended media, particularly at the 200 ppm rate (28 2.1 versus 39 2.2 mm, FIG. 6). Neither fungus grew on PDA amended with 500 ppm radicinin. In addition, when A. carotiincultae was grown on radicinin-amended media, the average number of transepta per conidium was re-

PRYOR AND GILBERTSON: TAXONOMIC STATUS OF ALTERNARIA CAROTIINCULTAE

55

FIG. 2. Colony morphology of A. radicina, A. carotiincultae, and A. petroselini after 15 d of growth on acidified potato dextrose agar (top row shows upper surface of the Petri dish, bottom row shows the lower surface).

duced and more subspherical conidia were produced compared with conidia produced on non-amended media (e.g., 3.6 septa/conidium at 200 ppm radicinin versus 4.9 septa/conidium for non-amended medium; FIGS. 7A, B). Similar results were obtained in three independent experiments.

Pathogenicity tests. Bartlett's test revealed homogeneity of variance within each experiment for each host. However, the chi-square test revealed non-homogeneity of variance among the three experiments. Thus, the trials were analyzed separately.

A. radicina and A. carotiincultae were highly pathogenic on carrot seedlings (TABLE IV) and both fungi had similar disease ratings. Neither taxon was pathogenic on parsley seedlings, and both were weakly or moderately pathogenic on celery, cilantro, and fennel seedlings (TABLE IV). In contrast, A. petroselini

FIG. 4. Microsclerotia produced by A. petroselini after 15 days of growth on V-8 agar. Scale bar represents 0.9 mm.

was highly to moderately pathogenic on celery, cilantro, fennel, and parsley seedlings, but was not pathogenic on carrot (TABLE IV). All isolates of all taxa were weakly or not pathogenic on caraway, dill, parsnip and anise seedlings (data not shown). None of the control seedlings developed disease symptoms.

Molecular analysis. Seventeen of the 20 random primers tested directed the amplification of reproducible RAPD fragments from total genomic DNA of all isolates tested, and a total of 710 markers were scored. Ten primers revealed polymorphisms between A. radicina and A. carotiincultae, whereas all 17 primers revealed polymorphisms between A. radicina and A. petroselini and between A. carotiincultae and A. petroselini. UPGMA cluster analysis of the RAPD data revealed 95% similarity between A. radicina and A. carotiincultae, a value considerably higher than those for the other fungal species examined

FIG. 3. Dendritic (tree-like) crystals of radicinin produced beneath a 15-day-old A. radicina colony growing on acidified potato dextrose agar.

FIG. 5. Thin-layer chromatography analysis of chloroform extracts prepared from culture filtrates of 10-day-old liquid cultures of A. radicina, A. carotiincultae, and A. petroselini. Lanes 1 and 2 contain 100 and 500 ppm radicinin, respectively. Lanes 3?6, A. radicina isolates 21-21-07, 21-2111, 21-21-14, and ATCC 6503. Lanes 7?9, A. carotiincultae isolates 21-21-13, 21-21-15, and 21-21-16. Lanes 10?12, A. petroselini isolates 21-41-01, 21-41-02, and 21-41-03.

56

MYCOLOGIA

FIG. 6. Colony morphology of (A) A. radicina isolate BMP 21?21?07, (B) A. carotiincultae isolate BMP 21?21?15, and (C) A. radicina isolate ATCC 6503 after 10 days of growth on potato dextrose agar amended with 0, 100, or 200 ppm radicinin.

(FIG. 8). For example, the species with the next highest similarity values were A. porri and A. solani (88%). A. petroselini was more similar to A. radicina and A. carotiincultae (84%) than it was to other Alternaria species (81%). RAPD analyses revealed few polymorphisms among the three A. radicina isolates, the three A. carotiincultae isolates, or the three A. petroselini isolates (FIG. 9).

Analysis of atypical A. radicina isolates. Conidium measurements as well as maximum and average number of transepta per conidium for ATCC 6503 and ATCC 58405 are presented in TABLE V, and are compared with those of the other A. radicina, A. carotiincultae, and A. petroselini isolates used in this study (TABLE I). The sizes of conidia of ATCC 6503 were nearly identical to those of A. radicina (TABLE V). The sizes of conidia of ATCC 58405 were similar but not identical to those of A. radicina, whereas the l/ w ratio and number of transepta were more similar to those of A. petroselini (TABLE V).

The growth of ATCC 6503 and ATCC 58405 on APDA after 8 d were 82 0.8 cm and 78 1.0 cm, respectively, which were considerably greater than those of A. radicina isolates and were comparable to those of A. carotiincultae and A. petroselini isolates, respectively (FIG. 1). On APDA, colony characteristics of ATCC 6503 included rapid growth and even margins; no production of crystals, pigments, or microsclerotia. These cultural characteristics are more similar to those of A. carotiincultae than those of A. radicina or A. petroselini. In addition, ATCC 6503 did

FIG. 7. Morphology of A. carotiincultae conidia after 10 days of growth on (A) PDA without radicinin and (B) PDA amended with 200 ppm radicinin. Scale bar represents 60 m.

not produce radicinin (FIG. 5). However, ATCC 6503 also produced few conidia in catenate arrangement, which is more similar to certain A. radicina isolates such as 21?21?14. On APDA, colony characteristics of ATCC 58405 included rapid growth and even margins, production of subspherical conidia and microsclerotia, but no production of crystals or catenate conidia and little production of yellow pigment. These cultural characteristics were more similar to those of A. petroselini than A. radicina or A. carotiincultae.

ATCC 6503 was pathogenic on carrot, weakly pathogenic on celery and cilantro, and non-pathogenic on parsley, dill, fennel, parsnip, caraway, and anise. In contrast, ATCC 58405 was highly pathogenic on celery, cilantro, fennel, and parsley, weakly pathogenic on caraway, carrot, and dill, and non-pathogenic on anise.

The RAPD patterns of ATCC 6503 were nearly identical to those of the A. carotiincultae isolates (FIG. 9). In contrast, the RAPD patterns of ATCC 58405

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download